DEVELOPMENTAL

BIOLOGY

143,111-121

(1991)

Isolation and Characterization PARKER

B. ANTIN

of an Avian Myogenic AND CHARLES

Accepted September

Cell Line

P. ORDAHL’

25, 19.90

Myogenic cell lines have proven extremely valuable for studying myogenesis i?c t&o. Although a number of mammalian muscle cell lines have been isolated, attempts to produce cell lines from other classes of animals have met with only limited success. We report here the isolation and characterization of seven avian myogenic cell lines (QMl-4 and QM6-8), derived from the quail fibrosarcoma cell line QT6. A differentiation incompetent QM cell derivative was also isolated (QM5,,). The major features of QM cell differentiation irr vitro closely resemble those of their mammalian counterparts. Mononucleated QM cells replicate in medium containing high concentrations of serum components. IJpon switching to medium containing low serum components, cells withdraw from the cell cycle and fuse to form elongated multinucleated myotubes. Cultures typically obtain fusion indices of 43-49s. Northern blot and immunoblot analyses demonstrate that each differentiated QM cell line expresses a wide variety of genes encoding muscle specific proteins: desmin, cardiac troponin T, skeletal troponin T, cardiac troponin C, skeletal troponin I, ru-tropomyosin, muscle creatine kinase, myosin light chain 2, and a ventricular isoform of myosin heavy chain. While all QM lines analyzed to date express at least some myosin light chain 2, only one line, QM7, expresses this gene at high levels. Surprisingly, none of the QM lines reported here express any known form of a-actin. The absence of sarcomeric actin expression may explain the absence of myofibrils in QM myotubes. These novel features of muscle gene expression in QM cells may prove useful for studying the role of specific muscle proteins during myogenesis. More importantly, however, the isolation of QM cell lines indicates that it may be feasible to isolate other avian mgogenic cells lines with general utility for the study of C 1991 Academic Press, Inc. muscle development. INTRODUCTION

Our current understanding of muscle development has been greatly aided by the discovery, over 30 years ago, that many essential aspects of myogenic differentiation could be reproduced in primary culture (Holtzer ef ul., 1958; Rinaldi, 1959; Konigsberg, 1960). Myoblast progenitor cells, isolated from embryonic or adult muscle tissue, proliferate in culture and then fuse into myotubes which undergo the molecular and cellular changes characteristic of terminal differentiation. Although such primary cultures have proven useful for studying many aspects of muscle differentiation, for certain types of experiments permanent clonal muscle lines offer distinct advantages. Yaffe and co-workers isolated the first permanent skeletal muscle cell lines over 20 years ago (Yaffe, 1968; Richler and Yaffe, 1970), and since then many other mammalian myogenic lines have been isolated and characterized (Schubert et al., 1974; Yaffe and Saxel, 1977a,b; Nadal-Ginard, 1978; Linkhart et ul., 1981; Blau et al., 1983; Konieczny and Emerson, 1984; Nakamigawa et al., 1988; Merrill, 1989). Unlike primary muscle cultures, cell lines provide a homogeneous and renewable source of replicating myoblasts which can be induced to differentiate upon alteration of media

’ To whom correspondence

should be addressed

components. Individual replicating myoblasts from cell lines can also be cloned, permitting the isolation of mutants that fail to express particular muscle phenotypes (Black and Hall, 1985; Gordon and Hall, 1989), or that are temperature sensitive for separate steps in the differentiation process (Loomis et al., 1973; Nguyen et al., 1983; Ackhurst et al., 1988). Clones harboring stably integrated exogenous DNA sequences have helped define the cis and tram elements involved in cell- and stagespecific regulation of muscle gene promoters (Melloul ef al., 1984; Nude1 et al., 1985; Hickey et al., 1986; Minty ef a,Z.,1986; Miwa et (II., 1987; Muscat and Kedes, 1987; Daubas et al., 1988; Jaynes et al., 1988; Sternberg et ol., 1988; Buskin and Hauschka, 1989) and have provided an j?~ vitro system for isolating myogenic determination genes (Wright et al., 1989). Cell lines have also proven useful for investigating the relationship between the cell cycle and myogenesis (Nadal-Ginard, 1978; Compton and Konigsberg, 1988) and for defining extracellular signals which influence muscle differentiation in vitro (Linkhart et al., 1981; Pinset and Whalen, 1984; Spizz et al., 1986; Pinset et al., 1988; Schneider and Olson, 1988; Florini and Magri, 1989). The advantages afforded by muscle cell lines have been largely restricted to the study of mammalian myogenesis since attempts to produce myogenic cell lines from other classes of vertebrates have met with only limited success. The lack of suitable muscle cell lines

112

DEVELOPMENTAL BIOLOGY

from avian species has been particularly problematic given the extensive body of work on avian myogenesis in viva and in primary cultures. Here we report the isolation and characterization of a series of stable, permanent myogenic cell lines derived from quail, which we have designated QM (for Quail Muscle). The QM cell lines were derived from the preexisting quail fibrosarcoma cell line QT6 (Moscovici et al., 1977), one of a series of cell lines cultured from methylcholanthrene-induced pectoralis fibrosarcomas of the Japanese quail Coturnix coturniz japonica. In most respects, QM cells closely resemble their mammalian counterparts. Differentiation of QM cells is serum dependent; following removal of serum, cells withdraw from the cell cycle, fuse into myotubes, and express many muscle-specific genes. In other respects, however, the QM cell lines described here have unique attributes, such as the absence of sarcomeric actin, which may prove useful for studying actin gene expression and myofibrillogenesis. MATERIALS

AND

METHODS

VOLUME 143, 1991

QT6 serialpassage at high density + QTGM Subclone

QMl

QM2 QM3

QM4

QM5,, QM6

QM7 QM8

FIG. 1. Flow chart showing the derivation of QM cell lines from the parental QT6 line. Stocks of QT6 cells serially passaged at high density for several years showed a 50- to loo-fold higher myogenic potential than the original QT6 cell population and were designated QTGM. QTGMH,, a highly myogenic suhclone of QTGM, was used to generate QM clones 1-8 via two rounds of limiting dilution subcloning. QM5,, was selected as a differentiation incompetent clone (DI) because cells failed to fuse or to express muscle genes following switch to differentiation medium.

Cell Culture QM cell lines were derived from QT6, a quail fibrosarcoma cell line available from the American Type Culture Collection (CRL 1708; ATCC, Rockville, MD). Our stocks of QT6 were obtained from the laboratory of J. Michael Bishop via the Cell Culture Center at UCSF. QT6 cells were grown on uncoated loo-mm plastic tissue culture dishes in growth medium (Ml99 medium plus Earles balanced salt solution [GIBCO, Grand Island, NY] supplemented with 10% tryptose phosphate, 10% fetal calf serum [FCS], glutamine, 100 U/ml penicillin, and 100 pg/ml streptomyocin). These QT6 stocks, maintained at high density and passaged continually for several years, were subsequently determined to contain l-2% of cells with myogenic potential. These stocks were designated QTGM to distinguish them from QT6 cells recently available from ATCC (passage 74) which contain 0.05-0.1% myogenic cells based upon staining with antibodies against desmin or myosin heavy chain. QM cell lines were derived from QT6M stocks via serial subcloning (see Fig. 1). QTGM were plated at 1 X lo4 cells per loo-mm uncoated plastic tissue culture dish in differentiation medium. Two clones which showed high fusion indices, QTGMH, and QTGMH,, were isolated using cloning rings. QTGMH, was selected for subsequent subcloning by serial dilution in 96-well plates. Individual clones were selected, split into several dishes, and grown in differentiation medium. When cultures approached confluence, some dishes were switched to differentiation medium (Ml99 medium plus Earles balanced salt solution supplemented with 0.5% FCS, penicillin and streptomyocin) for 72 hr and then scored for

percentage nuclei in myotubes and expression of myosin heavy chain. Seven highly myogenic colonies were selected (QMl-4 and QM6-8), plus a clone which failed to form myotubes or to express myosin heavy chain in differentiation medium (QM5,,). Each was subcloned once again via limiting dilution. Individual QM clones were then maintained as replicating populations at subconfluent density in growth medium on uncoated plastic tissue culture dishes. To induce myogenic differentiation, cells were grown to 90% confluence and then switched to differentiation medium for 48-120 hr. Antibodies The rabbit antisera prepared against desmin (Miles Scientific, Naperville, IL) binds to intermediate filaments in skeletal, cardiac, and smooth muscle cells but not to the vimentin intermediate filaments of fibroblasts (Mar et al., 1988). Rabbit antisera prepared against chicken muscle creatine kinase immunoprecipitates a single 40-kDa polypeptide from in vitro translation reactions using Day 18 chick leg muscle RNA. Identification of immunoprecipitated polypeptide as muscle creatine kinase was confirmed by peptide mapping (Ordahl et ah, 1984). The two mouse monoclonal antibodies, C4 and HUC l-l, were the gift of Dr. J. Lessard (Children Hospital Research Foundation, Cincinnati, OH). Monoclonal antibody C4 was prepared against chicken gizzard actin and recognizes all known vertebrate isoactins (Lessard, 1988). Monoclonal antibody HUCl-1 was prepared against vascular smooth muscle actin and

ANTINAND~RDAHL

QM: An Avinn Myogenic

recognizes all mammalian and avian muscle actins, including a-skeletal, a-cardiac, a-vascular, and y-enteric actins (Sawtell et al., 1988). Monoclonal antibody MF20 was prepared against adult chicken pectoralis myosin and recognizes all known avian sarcomeric myosins (Bader et ah, 1982). MF20 cell supernatant was obtained from the Developmental Studies Hybridoma Bank maintained by the Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland, and the Department of Biology, University of Iowa, Iowa City, Iowa, under Contract NOl-HD-6-2915 from the NICHD. Monoclonal antibodies EB165,2E9, and AB8 recognize avian embryonic, neonatal fast, and adult fast skeletal myosin heavy chain isoforms, respectively (Cerny and Bandman, 1987). Monoclonal antibody HVll, prepared against adult chicken ventricular myosin heavy chain, binds to the adult ventrical and to new muscle fibers in the embryo, fetus, and regenerating adult muscle tissue. This antibody binds transiently to all differentiating myotubes in culture (Wick and Bandman, 1989). Monoclonal antibodies EB165, 2E9, ABB, and HVll were the gift of E. Bandman (University of California, Davis). Monoclonal antibody S46, the gift of F. Stockdale (Stanford University), recognizes the slow myosin heavy chain isoforms SMl and SM2 (Stockdale and Miller, 1987).

Immunoblots QM or chick embryo pectoralis breast muscle cultures were washed twice with ice-cold phosphate-buffered saline (PBS) and cells were scraped into ice-cold PBS containing 10 mMEDTA, 0.2 mMphenylmethylsulfony1 fluoride, and 10 pg/ml aprotinin as protease inhibitors (PBS-PI). Cells were pelleted and then sonicated in lOO200 ~1 of PBS-PI. Whole cell homogenates were mixed with 1 vol of 2~ Laemmli sample buffer (Laemmli, 1977) containing 10% 2-mercaptoethanol, boiled for 3 min, and clarified by centrifugation. Aliquots normalized to protein content were separated by SDS-PAGE (8 or 10% gels) and the separated proteins were transferred electrophoretically to nitrocellulose in a buffer consisting of 12.5 mMTris base, 96 mMglycine, 0.1% SDS, and 25% methanol. Filters were preincubated for 30 min in 5% nonfat dry milk, 0.01% Anti-foam A (Sigma Chemical Co., St. Louis, MO) in PBS, then washed briefly in PBS plus 0.05% Tween 20 (PBS-Tween). All subsequent incubations and washes were performed in PBS-Tween. Filters were incubated in the appropriate dilution of primary antisera or monoclonal antibodies, then with biotinylated horse anti-mouse or goat anti-rabbit secondary antibodies (Vector Laboratories, Burlinghame, CA). Bound antibodies were visualized using the avidin-

113

Cdl Line

biotin-alkaline phosphatase kit, Vector Laboratories).

system (Vectastain

ABC

RNA Blots Total RNA was isolated from QM and cultured quail pectoralis muscle cells according to the method of Chomcyznski and Sacchi (1987). Northern analysis of QM mRNAs was performed essentially as described (Maniatis et al., 1982). Characterization of cDNA probes used for Northern analyses are as follows: Cardiac troponin T, 622nt cDNA fragment (Cooper and Ordahl, 1985); skeletal (fast) troponin I, 497nt cDNA fragment (Nikovits et al., 1986); cardiac/slow troponin C, 420 nt cDNA fragment (clone pcCll1 of Maisonpierre et al., 1987); skeletal (fast) troponin T, 560 nt cDNA fragment (clone pcC119 of Hastings et ah, 1985); skeletal (fast) a-tropomyosin, 985 nt cDNA fragment (clone pTM4 of Macleod, 1982); muscle creatine kinase, 390 nt cDNA fragment (clone pMCK-X3 of Ordahl et al., 1984); myosin heavy chain (skeletal fast), 1lOnt cDNA fragment (clone pMHCl-X; C. P. Ordahl, unpublished sequence which matches nucleotides 1110-1121 of sequence published by Kavinsky et al, 1983); myosin light chain 2,113 nt cDNA fragment with a nt sequence corresponding to amino acids No. 68-111 of chicken myosin light chain 2 published by Matsuda et al., 1977; skeletal a-a&in, 3’ untranslated region of chicken skeletal a-actin (clone paactin 1 of Ordahl et ab, 1980); and total actin; a chicken skeletal oc-actin coding region cDNA fragment (corresponds to clone pa-actin 2 of Ordahl et al., 1980). Southern Analysis Genomic DNA was isolated from newborn mouse liver, Day 8 whole quail and chicken embryos, and cultured QM7 cells essentially as described (Ordahl, 1977). Southern analysis of DNA was performed as described (Church and Gilbert, 1984), using a 3’ untranslated region skeletal n-a&in probe (pa-a&in 1, above). RESULTS

Derivation

of Quail Myogenic Cell Lines fwrn

QT6

The QT6 cell line was originally characterized as highly transformed, based upon a rounded morphology and poor adherence to substrate, the ability to produce tumors in host birds and form colonies in soft agar, and an enhanced rate of hexose uptake. The QT6 cells used for this study, which had been repeatedly passaged at high density for several years, lost many of these transformed characteristics. Unlike the rounded transformed phenotype of QT6 cells freshly obtained from the ATCC, our stocks of QT6 were flattened on the substrate and fibroblastic in overall morphology.

114

DEVELOPMENTALBIOLOGY VOLUME143.1991

Several initial observations led to the conclusion that our laboratory QT6 stocks contained cells with myogenie potential. First, when QT6 cells were grown to high density in nutrient-rich growth medium, occasional multinucleated structures containing 3 to 10 nuclei were observed. Second, some neomycin-resistant clones harboring recombinant chicken cardiac troponin T minigenes also expressed full-length endogenous (quail) cardiac troponin T mRNAs when grown to high density (T. A. Cooper, unpublished observation). Third, if clones expressing endogenous cardiac troponin T mRNA were shifted to differentiation medium (low serum), up to 20% of cells fused to form multinucleated structures. These multinucleated cellular structures bound antibodies against muscle myosin heavy chain. Our laboratory QT6 stocks containing cells with myogenie potential were subsequently designated QTGM, to distinguish them from parental QT6 available from ATCC. To determine whether permanent, clonal myogenic cell lines could be derived from QTGM, cells were plated at low density in growth medium and individual clones were isolated (Fig. 1). One clone, designated QTGMH,, showed a high fusion index and was picked for subcloning by serial dilution. Subclones were screened for the percentage of nuclei within myotubes and expression of muscle myosin heavy chain following switch to differentiation medium. Eight clones were isolated and designated QMl-8. Subcloning of the QM clones was repeated and individual clones were selected based upon growth index in growth medium and the formation of myotubes and expression of myosin heavy chain in differentiation medium. One clone, QM5,,, was designated “differentiation incompetent” because it failed to fuse into myotubes or to synthesize myosin heavy chain following switch to differentiation medium. Morphology

of Diflerentiated

QM Cultures

Differentiation competent QM clones maintained in growth medium consisted of a population of rapidly dividing mononucleated cells (Fig. 2A). Upon growth to high density and switch to differentiation medium, cells ceased dividing and during the next 72 hr up to 50% fused to form highly elongated multinucleated myotubes (Figs. 2B-2F). Immunocytochemical staining of typical differentiated QM7 cultures using antibodies against myosin heavy chain (Figs. 2C and 2E) or desmin (Figs. 2D and 2F) illustrates the large numbers of elongated branching myotubes that form in these cultures. Within each culture, l-2% of mononucleated QM7 cells also bound anti-desmin and anti-myosin heavy chain (Figs. 2E and 2F). Staining with anti-myosin heavy chain showed a diffuse nonfibrillar distribution within

myotubes. A sarcomeric banding pattern reflecting the presence of organized myofibrils was never observed. These results indicate that QM cells have the general characteristics of a myogenic cell line. Veri&cation

that QM Cells Are Derived from Quail

To verify the species identity of QM cells, a chicken skeletal a-actin-specific cDNA probe (pa-a&in 1, Ordahl et al., 1980) was hybridized to restriction digests of mouse, chicken, quail, and QM7 DNA (Fig. 3). The bands observed in restriction digests of chicken DNA (lanes 3 and 4) correspond to those previously reported (Ordahl et al., 1980). The hybridization patterns of this probe to quail DNA (lanes 5 and 6) and QM7 DNA (lanes 7 and 8) were essentially identical to each other and distinct from that of either mouse (lanes 1 and 2) or chicken (lanes 3 and 4). Further confirmation that QM cells are of quail origin was obtained by S-l analysis of QM RNA. A chicken cardiac troponin T probe from the 5’ half of the cDNA protected a full-length fragment of 618 nucleotides (T. Cooper, data not shown). Nuclease protection in this region would not be expected if QM cells were of mammalian origin since comparison of the chicken and rat cTNT sequences shows only limited homology over this region (Jin and Lin, 1989; Cooper and Ordahl, 1985). Time Course of QM Cell Differentiation A time course of cell fusion and myosin heavy chain expression was determined for QM7 cultures following switch to differentiation medium. Cultures grown to high density in growth medium were switched to differentiation medium, and at various intervals thereafter the percentage of nuclei in myotubes and the percentage of nuclei in myosin heavy chain-positive cytoplasm were determined. Results of a typical time course experiment are shown in Fig. 4. Immediately before switching to differentiation medium (time zero), less than 2% of cells had fused to form myotubes. After medium switch the rate of fusion increased rapidly and by 72 hr 46% of cells had fused into myotubes. As indicated by the slightly higher percentage of nuclei within the cytoplasm expressing myosin heavy chain (47%) and by the immunocytochemical data presented above (Figs. 2E and 2F), a small but constant percentage (l-2%) of QM7 cells expressed myosin heavy chain but remained mononucleated. Maximum fusion indices for other differentiation competent QM clones after switch to differentiation medium ranged from 43 to 49% (data not shown). QM7 cells maintained in growth medium continued to replicate, and Fig. 4 shows that only after several days at confluent density did a small percentage of cells fuse

ANTINAND ORDAHL

QM: An Avian

Myogenic

Cell Line

115

FIG. 2 (A and B) Phase micrographs of QM7 cells in growth medium (A) and 72 hr following switch to differentiation medium (B). QM cells in growth medium are flattened on the substrate and replicate with a doubling time of between 12 and 16 hr. Bars, 100 pm. (C-F) Bright-field micrographs of QM7 cultures ‘72 hr following shift to differentiation medium. Alkaline phosphatase reaction product shows binding of antibody MF20, against myosin heavy chain (C, E), or anti-desmin (D, F). (E and F) Higher magnification views of (C) and (D), respectively. Large numbers of elongated branching myotubes are subtended by a confluent layer of mononucleated cells, the vast majority of which do not bind either antibody. Note in (E) and (F) the occasional mononucleated cells which stain with anti-desmin and anti-myosin (arrows). Bars; (C, D) 1 mm; (E, F) 200 pm.

116

DEVELOPMENTAL BIOLOGY

12345678

VOLUME 143.1991

cell fusion (Fig. 4).

following

shift

to differentiation

medium

QM Myotubes Express Many Muscle-SpeciJic Genes

FIG. 3. Southern analysis of genomic DNA from mouse (lanes 1 and Z), chicken (lanes 3 and 4), quail (lanes 5 and 6), and QM7 (lanes 7 and 8). DNAs were digested with Hind111 (lanes 1, 3, 5, and 7) or with BanzHI (lanes 2,4,6, and 8). Electrophoresed DNAs were transferred to nitrocellulose and hybridized with a cDNA probe encoding the 3’ untranslated region of chicken skeletal a-actin. This probe binds to a smear in lanes containing mouse DNA (lanes 1 and 2) and to specific bands in lanes containing chick (lanes 3 and 4), quail (lanes 5 and 6), or QM7 (lanes ‘7 and 8) DNA. The hybridization patterns of quail and QM7 DNAs appear identical. Arrows indicate identical bands in BarnHI-digested quail and QM7 DNA (lanes 6 and 8).

into myotubes (8%) and express myosin heavy chain (10%). Cultures maintained in growth medium at subconfluent density contained a small fraction (l-2%) of differentiated muscle cells that expressed both myosin heavy chain and desmin (data not shown), although the vast majority of cells continued to cycle and did not express detectable levels of either of these muscle-specific proteins. To determine the time course of muscle mRNA accumulation in QM cells, total RNA was isolated from QM2 cultures at 24-hr intervals following switch to differentiation medium and was analyzed by Northern analysis using cDNA probes encoding cardiac troponin T and skeletal troponin I. Immediately prior to medium switch, mRNAs encoding cardiac troponin T and skeletal troponin I were present at low but detectable levels (Figs. 5A and 5B, lanes 1). This low level expression may be due to the l-2% of differentiated muscle cells present in cultures maintained in growth medium (see above). Levels of cardiac troponin T and skeletal troponin I mRNAs increased rapidly following switch to differentiation medium (Fig. 5A, lanes 2-4, Fig. 5B, lanes 2-5). Thus, the time course of accumulation of muscle-specific protein and mRNA correlates with the kinetics of

Whole cell homogenates of QM7 cells were also analyzed by immunoblot using an antiserum prepared against muscle creatine kinase. Muscle creatine kinase was not detectable in homogenates of QM7 cells maintained in growth medium at subconfluent density (Fig. 6, lane 1). In contrast, anti-muscle creatine kinase bound strongly to a band in lysates from QM7 cultures switched to differentiation medium for 96 hr (Fig. 6, lane 2). To assess the expression of muscle-specific mRNAs in the QM cell lines upon switch to differentiation medium, total RNA was isolated from QM cultures following 96 hr in differentiation medium and was assayed by Northern analysis using cDNA probes coding for several contractile protein mRNAs. RNA from Day 10 quail embryo pectoralis muscle was included as a control. Figure 7 shows that each differentiation-competent QM

24

48

72

96

Hours FIG. 4. Time course of cell fusion and myosin heavy chain expression by QM7 cells following switch to differentiation medium. QM7 cultures were grown almost to confluence in growth medium and then were either maintained in growth medium or switched to differentiation medium. Cultures were fixed at various times and stained with the monoclonal antibody MF20 (anti-myosin heavy chain) and the nuclear binding dye DAPI. Cultures were scored for the percentage of nuclei within myotubes (three or more nuclei within a common cytoplasm) and for the percentage of nuclei within myosin heavy chainpositive cells. Cultures maintained continually in growth medium: closed circles, percentage of nuclei in myotubes; open circles, percentage of nuclei within myosin heavy chain-positive cells. Cultures switched to differentiation medium at zero hours: closed squares, percentage of nuclei in myotubes; open squares, percentage of nuclei within myosin heavy chain positive cells. The difference between the maximum fusion index (45% at 72 hr) and the percentage of nuclei within myosin heavy chain-positive cells (47% at 72 hr) represents the percentage of myosin heavy chain-positive cells which failed to fuse.

117 A

B 12345

1234

Qhl# Lane

a

345678 b c

d

e

f

g

sTNT

cTNC

sTNl

MLC* MHC fast skel

FIG. 5. Time course analysis of myofibrillar mRNA accumulation following shift of QM2 cells to differentiation medium. (A) Northern blot probed with chicken cardiac troponin T cDNA. Lanes l-4 contain mRNAs from cultures 0,24,60, and 96 hr following switch to differentiation medium. (B) Northern blot probed with skeletal (fast) troponin I cDNA. Lanes l-5 contain mRNAs from cultures 0, 24, 48, ‘72, and 96 hr following switch to differentiation medium.

clone accumulates mRNAs coding for cardiac troponin T, skeletal (fast) troponin T, cardiac/slow troponin C, skeletal (fast) troponin I, a-tropomyosin (fast), and muscle creatine kinase. QM5,, cells failed to fuse in differentiation medium and did not accumulate any of these muscle-specific mRNAs (Fig. 7, lane d). Absolute mRNA levels varied from clone to clone; QM7 muscle

12

L,.

- 42

FIG. 6. Immunoblot analysis of muscle creatine kinase expression in QM7 cultures. Whole cell homogenates from QM7 cells in growth medium (lane 1) and 96 hr following switch to differentiation medium (lane 2) were blotted and probed using anti-muscle creatine kinase serum. Muscle creatine kinase is detectable only in QM? cultures following switch to differentiation medium (lane 2).

FIG. ‘7. Northern analysis of total RNA isolated from QM clones 96 hr following switch to differentiation medium. RNAs were probed with cDNAs specific for cardiac troponin T (cTNT); skeletal (fast) troponin T (sTNT); cardiac/slow troponin C (cTNC); skeletal (fast) troponin I (sTN1); Lu-tropomyosin (fast) (cuTM); muscle creatine kinase (MCK); myosin light chain 2 (ML&); and skeletal (fast) myosin heavy chain. QM No. corresponds to the numbered designation of the individual QM clones.

cells expressed particularly high levels of each mRNA (Fig. 7, lane f), comparable to levels observed for quail breast muscle (lane a). Interestingly, a cDNA probe encoding a portion of myosin light chain 2 hybridized strongly only with control RNA isolated from primary pectoralis cultures and with RNA from QM7 cultures (Fig. ‘7, lanes a and f), although a prolonged exposure of the blot shown in Fig. 7 revealed low levels of myosin light chain 2 mRNA in all differentiation-competent QM clones (data not shown). Although embryonic quail pectoralis accumulate mRNAs which hybridize with a probe encoding part of a fast skeletal isoform of myosin heavy chain (Fig. 7, lane a), RNAs from the QM clones fail to bind this probe, indicating that QM cells express a myosin heavy chain isoform that is distinct from the fast skeletal isoform (see below). Overall, these results indicate that each of the differentiation-competent QM cell lines expresses a wide variety of muscle-specific genes encoding thin and thick filament sarcomeric proteins.

Differentiated QM muscle cells readily bound monoclonal antibody MF20 (Figs. 2C and 2E), which recognizes all known chicken myosin heavy chain isoforms (Bader et al., 1982). However, a llO-nt cDNA fragment of a chicken fast skeletal myosin heavy chain failed to

118

DEVELOPMENTAL BIOLOGY

FIG. 8. Immunoblot analysis of myosin heavy chain isoform expression in QM cells. Whole cell homogenates of QM7 cells 96 hr following switch to differentiation medium were separated by SDS-PAGE (8% gels) and transferred to nitrocellulose. Blots were challenged using monoclonal antibodies with the following reactivities: MF20, pan reactive with all known avian myosin heavy chain isoforms; EB165, embryonic skeletal myosin heavy chain; 2E9, neonatal fast skeletal myosin heavy chain; AB8, adult fast skeletal myosin heavy chain; HVll, adult ventricular myosin heavy chain; S46, SMl and SM2 slow myosin heavy chain isoforms. Only MF20 and HVll bind to a band in QM lysates. Doublet observed is due to partial degradation of myosin heavy chain.

hybridize to QM RNAs (Fig. 7), suggesting that QM cells express a myosin heavy chain isoform not closely related to the fast skeletal isoform. To identify the myosin heavy chain isoform(s) synthesized by differentiated QM cells, whole cell homogenates of QM7 were analyzed by immunoblot using a panel of isoform-specific monoclonal antibodies. Figure 8 shows that while MF20 readily bound to a band in QM7 lysates, monoclonal antibodies directed against embryonic (EB165), neonatal fast (2E9), and adult fast (AB8) skeletal muscle myosin heavy chain (Cerny and Bandman, 1987) did not recognize this band. Monoclonal antibody S46, which recognizes the slow SMl and SM2 myosin heavy chain isoforms, also failed to bind to a band in QM7 lysates. However, monoclonal antibody HVll, which recognizes a myosin heavy chain isoform present in the adult ventrical and transiently in all new and regenerating muscle fibers (Wick and Bandman, 1989), bound to a band of the same mobility as that recognized by MF20. Thus, QM cells express a myosin heavy chain isoform that is closely related, or identical, to the ventricular myosin heavy chain. QM Muscle Cells Do Not Express a-A&n Surprisingly, QM clones grown in differentiation medium did not accumulate mRNAs coding for a-actin (Fig. 9). A cDNA probe from the coding region of skeletal a-a&in, which cross-hybridizes with mRNAs coding for all actin isoforms (Ordahl et al., 1980), hybridized to mRNAs encoding the nonmuscle fi- and y-a&ins in differentiated QM cultures, but showed no detectable hybridization to mRNAs of the expected size for skeletal or cardiac a-a&in (Fig. 9A). The absence of skeletal a-actin mRNA was confirmed using pa-a&in 1, a 3’ untrans-

VOLUME 143,199l

lated cDNA probe that is specific for this iso-mRNA (Fig. 9B). Immunoblot analysis confirmed the absence of a-actin isoproteins in QM muscle cells. Monoclonal antibody C-4, which binds to all known vertebrate actins (Lessard, 198X), bound to a prominent band in whole cell lysates prepared from cultured quail pectoralis muscle and from cultures of QM7 cells following 96 hr in differentiation medium (Fig. 10, lanes 1 and 2). However, monoclonal antibody HUCl-1, which binds selectively to muscle iso-actins (skeletal, cardiac, and vascular (u-actins and y-enteric actin [Sawtell et ah, 19SS]) failed to bind any protein in the QM7 muscle homogenates (Fig. 10, lane 4) but bound to a prominent band in homogenates of cultured quail pectoralis (Fig. 10, lane 3). Therefore, differentiated QM cultures express nonmuscle cytoplasmic actins but apparently do not express any known form of sarcomeric actin. This deficiency may explain the absence of striated myofibrils in these cells (see Discussion). DISCUSSION

We have isolated and characterized permanent clonal skeletal muscle cell lines (designated QM) from QT6, an established quail fibroblastic cell line. Similar to the numerous established mammalian muscle cell lines already described, differentiation of QM cell lines is serum dependent; cells divide rapidly in nutrient-rich medium and upon removal of serum, a high percentage of cells cease dividing and fuse to form multinucleated

QM# Lane

a

345676 b c

d

e

f

g

-a

FIG. 9. Actin mRNA expression in QM cell lines. Northern blots of total RNA isolated from Day 10 quail embryo pectoralis (lane a) or QM cultures 96 hr following shift to differentiation medium (lanes b-g). Blots were probed with pa-actin 2 (A), which recognizes all cellular actin mRNAs, or with pa-actin 1 (B), which is specific for the 3’ untranslated region of the chicken skeletal a-actin mRNA. The expected positions of the mRNAs encoding p-, y-, and a-type RNAs are indicated. Note that no hybridization is detected at the position expetted for a-type actin mRNA with either probe.

ANTIN

AND ORDAHL

transformed founder cells. Based upon the age of the quail at the time of carcinogen injection (l-4 weeks), it is therefore possible that the initially transformed cell(s) may have included a muscle satellite cell. Immunofluorescence analysis of fresh stocks of QT6 from ATCC using anti-myosin heavy chain and anti-desmin showed that 0.05-0.1% of cells expressed at least one of these muscle markers (data not shown), although such cells were rounded and only loosely adherent to the substrate. Thus, even in QT6 samples which closely resembled the original isolates described by Moscovici et al. (1977), at least a small percentage of cells expressed muscle-specific proteins. Long-term maintenance of ! QT6 cells at high density in our hands may have been I 5. preferentially enriched for these myogenic cells, giving rise to the mixed line we have designated QT6M (see Fig. 1). While we currently favor the second hypothesis, adanalysis FIG. 10. Actin protein expression in QM7 cells. Immunoblot of whole cell homogenates from cultured quail pectoralis muscle ditional experiments will be required to discriminate (lanes 1 and 3) or QM7 cultures 96 hr following switch to differentiabetween these two possibilities. tion medium (lanes 2 and 4). Monoclonal antibody C-4, which recogUnexpectedly, none of the QM clones described here nizes all known vertebrate actins, binds strongly to a band in homogeform cross-striated myofibrils. The failure to assemble nates from both quail pectoralis and QM7 cultures (lanes 1 and 2). myofibrils is likely related to the absence of any cu-actin Monoclonal antibody HUCl-1, which binds selectively to skeletal, cardiac, and vascular n-actins and y-enteric actin, binds strongly to exisoproteins in these cells. Cultured embryonic chicken tracts from quail pectoralis cultures (lane 3), but fails to bind to expectoralis expresses the cardiac and skeletal cu-actin tracts of differentiated QM7 cultures (lane 4). isoforms (Hayward and Schwartz, 1986), and a similar pattern of expression might be expected in QM muscle cells. Comparison of actin mRNA levels between the difmyotubes. Although growth to high density prior to me- ferentiation incompetent QM5,, (Fig. 9, lane d) and the dium shift leads to the formation of large numbers of differentiation competent QM clones (Fig. 9, lanes b, c, f, elongated myotubes, neither high density nor fusion is a and g) indicates that expression of p- and y-actin remains approximately constant in QM cells both prior to prerequisite for differentiation of QM cells. Differentiated QM cells express many muscle-specific genes, in- and following differentiation. The absence of cu-actin therefore did not lead to a compensatory up-regulation cluding desmin, muscle creatine kinase, myosin light chain 2, and genes encoding a full complement of the of these nonmuscle actin isoforms nor did the cytoskelemajor thin filament-associated proteins; troponin T, tal forms of actin substitute for a-actin in the assembly of striated myofibrils. However, the cytoplasmic levels troponin I, troponin C, and a-tropomyosin. Interestingly, the only form of myosin heavy chain detected in of /3- and y-actins, or of other myofibrillar proteins, may QM cells bound an antibody prepared against a cardiac be insufficient to allow myofibril assembly to occur. It ventricular myosin heavy chain. QM cells do not express will be interesting to determine whether the introduction of u-actin into QM myotubes will lead to myofibril any form of sarcomeric actin, which may explain their assembly. inability to assemble striated myofibrils (see below). The nature of the lesion that inhibits expression of all The question of how QM cells arose within the QT6 line is an intriguing one. One possibility is that a sub- isoforms of cu-actin is not yet known. Although it is posfraction of immortal QT6 cells underwent a secondary sible that one or more of the unlinked cr-actin genes transformation event that activated one of the muscle could be missing, Southern blot analysis has shown that determination genes, such as MyoD, (Davis et ab, 1987), the skeletal cY-actin genes are grossly present. It would giving rise to a secondary myogenic lineage. On the seem more likely that QM cells have a regulatory defect other hand, it is possible that the QT6 line has always which precludes transcription of the a-actin genes or, leads to rapid degradation of a-actin contained a small fraction of myogenic cells. The paren- alternatively, tal QT6 cell line was originally established from a quail mRNAs. It will be interesting to determine if QM muspectoralis tumor placed in mass culture following tryp- cle cells contain CArG box-binding proteins, which are sinization (Moscovici et ah, 1977). Since the original QT6 required for cy-actin gene transcription (Miwa et al., cultures were not clonally derived, it is not known 1987). Differentiated QM7 cultures contain highly elongated whether QT6 arose from the progeny of one or more

12

34

120

DEVELOPMENTALBIOLOGY

bipolar myotubes which can course for several hundred microns along the substrate (Figs. 2C and 2D). Failure of QM myotubes to assemble a contractile apparatus clearly has little effect on their ability to initiate or maintain a bipolar morphology. Although the mechanisms which govern such highly specialized cell morphologies are not yet understood, the generation and maintenance of bipolarity probably involves a complex interplay among cytoskeletal elements, membrane ligands, and extracellular matrix components. In this regard it is interesting that matrix substrates such as rat tail collagen, gelatin, and laminin are not required for myotube formation in QM cultures and actually lead to a reduction in number and size of myotubes which form (data not shown). Immunofluorescence analysis revealed that QM myotubes contain small numbers of pand y-actin-containing stress fibers localized almost exclusively at their growth tips (unpublished observations). These terminal stress fibers may interact with extracellular matrix components synthesized by QM cells to generate and maintain myotube cell shape. Alternatively, experiments using microtubule disrupting and stabilizing agents in primary muscle cultures suggest that microtubules play a prominent role in the formation and maintenance of bipolarity in muscle cells (Antin et al., 1981; Toyama et al., 1982). QM cells may provide a system for investigating the contributions of nonmuscle contractile proteins, microtubules, and extracellular matrix components to the processes of myotube morphogenesis and myofibril formation. We thank Thomas Cooper for sharing his original observations on the expression of the endogenous cardiac troponin T gene in our QT6 stocks. We also thank our colleagues for making probes and cells available to us, and Nina Kostanian for expert technical assistance. This work was supported by USDA Grant %‘CRCR-12404 and by NIH Grants GM32018 and HL35561. REFERENCES ACKHURST,R. J., FLAVIN, N. B., and WORDEN,J. (1988). Isolation and characterization of a variant myoblast cell line that is temperature sensitive for differentiation. Mol. Cell. Biol. 8,2335-2341. ANTIN, P. B., FORRY-SCHAUDIES, S., FRIEDMAN,T. M., TAPSCOTT,S. J., and HOLTZER,H. (1981). Taxol induces postmitotic myoblasts to assemble interdigitating microtubule-myosin arrays that exclude actin filaments. J. Cell Biol. 89, 300-308. ANTIN, P. B., KARP, G. C., and ORDAHL,C. P. (1991). Transgene expression in the quail muscle myogenic cell line. Dev. Biol. 143. BADER,D., MASAKI, T., and FISCHMAN,D. A. (1982). Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro. J. Cell Biol 95, 763-770. BLACK, R. A., and HALL, Z. W. (1985). Use of a replica technique to isolate muscle cell lines defective in expressing the acetylcholine receptor. Proc. N&l. Acad. Sci. USA K&124-128. BLAU, H., CHIU, C-P., and WEBSTER,C. (1983). Cytoplasmic activation of human nuclear genes in stable heterokaryons. Cell 32,1171-1180. BUSKIN,J. N., and HAUSCHKA,S. D. (1989). Identification of a myocyte

VOLUME143,1991

nuclear factor that binds to the muscle-specific enhancer of the mouse muscle creatine kinase gene. Mol. Cell. BioL 9,2627-2640. CERNY,L. C., and BANDMAN, E. (1987). Expression of myosin heavy chain isoforms in regenerating myotubes of innervated and denervated pectoral muscle. Dev. Biol. 119,350-362. CHOMCZYNSKI,P., and SACCHI,N. (1987). Single step method of RNA isolation by acid guanidinium thioeyanate-phenol-chloroform extraction Anal. B&hem. 162,156-159. CHURCH,G. M., and GILBERT, W. (1984). Genomic sequencing. Proc. Nat1 Acad. Sci. USA 81,1991-1995. COMPTON,R. S., and KONIGSBERG,I. R. (1988). Cell cycle withdrawal without concomitant differentiation: Analysis of a Gl-specific temperature-sensitive murine myoblast cell line. Dev. Biol. 129, 476494.

COOPER,T. A., and ORDAHL,C. P. (1985). A single cardiac troponin T gene generates embryonic and adult isoforms via developmentally regulated alternate splicing. J. Biol. Chem. 260, 11,140-11,148. DAUBAS, P., KLARSFIELD, A., GARNER, I., PINSET, C., Cox, R., and BUCKINGHAM,M. (1988). Functional activity of the two promoters of the myosin alkalai light chain gene in primary muscle cell cultures: Comparison with other muscle gene promoters and other culture systems. Nucleic Acids Res. 16,1251-1271. DAVIS, R. L., WEINTRAUB,H., and LASSAR,A. B. (1987). Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cell 51, 987-1000. FLORINI, J. R., and MAGRI, K. A. (1989). Effects of growth factors on myogenic differentiation. Amer. J. Physiol 265, C701-‘711. GORDON,H., and HALL, Z. W. (1989). Glycosaminoglycan variants in the C2 muscle cell line. Dev. Biol. 135, 1-11. HASTINGS,K. E., BUCHER,E. A., and EMERSON,C. P. (1985). Generation of troponin T isoforms by alternative RNA splicing in avian skeletal muscle. J. Biol. Chem. 260, 13,699-13,703. HAYWARD,L. J., and SCHWARTZ,R. J. (1986). Sequential expression of chicken actin genes during myogenesis. J. Cell Biol. 102,1485-1493. HICKEY, R., SKOULTCHI,A., GUNNING,P., and KEDES,L. (1986). Regulation of a human cardiac actin gene introduced into rat L6 myoblasts suggests a defect in their myogenic program. Mol. Cell. BioL 6,32873290.

HOLTZER,H., ABBOT,J., and LASH, J. (1958). On the formation of multinucleated myotubes. Armt. Rec. 131, 567a. JAYNES, J. B., JOHNSON,J. E., BUSKIN, J. N., GARTSIDE, C. L., and HAUSCHKA, S. D. (1988). The muscle creatine kinase gene is regulated by multiple upstream elements, including a muscle-specific enhancer. Mol. Cell. BioL 8, 62-70. JIN, J. P., and LIN, J. J.-C. (1989). Isolation and characterization of cDNA clones encoding embryonic and adult isoforms of rat cardiac troponin T. J. Biol. Chem. 264,14,771-14,777. KAVINSKY, C. J., UMEDA, P. K., SINHA, A. M., ELZINGA, M., TONG, S. W., ZAK, R., JAKOVCIC,S., and RABINOWITZ,M. (1983). Cloned mRNA sequences for 2 types of embryonic myosin heavy chains from chick skeletal muscle. I. DNA and derived amino acid sequence of light meromyosin. J. Biol. Chem. 258,5196-5205. KONIECZNY,S. F., and EMERSON,C. P. (1984). 5-Azacytidine induction of stable mesodermal stem cell lineages from lOT1/2 cells: Evidence for regulatory genes controlling determination. Cell 38,791-800. KONIGSBERG,I. R. (1960). The differentiation of cross-striated myofibrils in short term cell culture. Exp. Cell Res. 21,414-420. LAEMMLI, IT. K. (1977). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Londm) 277,680685. LESSARD,J. L. (1988). Two monoclonal antibodies to actin: One muscle selective and one generally reactive. Cell Motil. Cytoskel 10,349-362. LINKHART, T. A., CLEGG,C. H., and HAUSCHKA,S. D. (1981). Myoblast differentiation in permanent clonal mouse myoblast cell lines: Regu-

ANTIN AND ORDAHL

QM: An Avian Myogenic

lation by macromolecular growth factors in the culture medium. Del?. Biol. 86, 19-30. LOOMIS, W. F., WAHRMANN, J. P., and LUZZATI, D. (1973). Temperature-sensitive variants of an established myoblast line. Froc. Natl. Acad. Sci. USA 70,425-429. MACLEOD, A. R. (1982). Distinct alpha-tropomyosin mRNA sequences in chicken skeletal muscle. Eur. J. Biochem. 126, 293-297. MAISONPIERRE, P. C., HASTINGS, K. E., and EMERSON, C. P., JR. (1987). The cloning and codon and amino acid sequence of the quail slow/ cardiac troponin C cDNA. 1r~ “Methods in Enzymology” (A. R. Means and P. M. Conn, Eds.), Vol. 139, pp. 326-337. Academic Press, New York. MANIATIS, T., FRITSCH, E. F., and SAMBROOK, J. (1982). “Molecular Cloning, a Laboratory Manual.” Cold Spring Harbor Press, Cold Spring Harbor, NY. MAR, J. H., ANTIN, P. B., COOPER, T. A., and ORDAHL, C. P. (1988). Analysis of the upstream regions governing expression of the chicken cardiac troponin T gene in embryonic cardiac and skeletal muscle cells. J. Cell Biol. 107, 573-585. MATSUDA, G., SUZUYAMA, Y., MAITA, T., and UMEGANE, T. (1977). The L-2 light chain of chicken skeletal muscle myosin. FEBS Lett. 84, 53356. MELLOUL, D., ALONI, B., CALVO, J., YAFFE, D., and NUDEL, U. (1984). Developmentally regulated expression of chimeric genes containing muscle actin DNA sequences in transfected myogenic cells. EMBO .I 3, 983-990. MERRILL, G. F. (1989). Clonai derivation of a rat muscle cell strain that forms contraction-competent myotubes. In Vitro Cell. Dev. Biol. 25, 471-476. MINTY, A., BLAU, H., and KEDES, L. (1986). Two-level regulation of cardiac actin gene transcription: Muscle-specific modulating factors can accumulate before gene activation. Mol. Cell. Biol. 6, 2137 2148. MIWA, T., BOXER, L. M., and KEDES, L. (1987). CArG boxes in the human cardiac alpha-actin gene are core binding sites for positive trans-acting regulatory factors. Proc. Natl. Acad. Ski. USA 84,67026706. MOSCOVICI, C., MOSCOVICI, M. G., JIMENEZ, H., LAI, M. M. C., HAYMAN, M. J., and VOGT, P. (1977). Continuous tissue culture cell lines derived from chemically induced tumors of Japanese quail. Cell 11, 95-103. MUSCAT, G. E., and KEDES, L. (1987). Multiple 5’-flanking regions of the human alpha-skeletal actin gene synergistically modulate muscle-specific expression. Mol. Cell. Biol. 7, 4089-4099. NADAL-GINARD, B. (1978). Commitment, fusion and biochemical differentiation of a myogenic cell line in the absence of DNA synthesis. Cell 15, 855-864. NAKAMIGAWA, T., MOMOI, M. Y., MOMOI, T., and YANAGISAWA, M. (1988). Generation of human myogenic cell lines by the transformation of primary culture with origin-defective SV40 DNA. J. Neural. sci. 83, 305-319. NGUYEN, H. T., MEDFORD, R. M., and NADAL-GINARD, B. (1983). Reversibility of muscle differentiation in the absence of commitment: Analysis of a myogenic cell line temperature-sensitive for commitment. Cell 34, 281-293. NIKOVITS, W. M., KUNCIO, G., and ORDAHL, C. P. (1986). The chicken fast skeletal troponin I gene: Exon organization and sequence. A%cleic Acids Res. 14, 3377-3390. NUDEL, I!., GREENBERG, D., ORDAHL, C. P., SAXEL, O., NEUMAN, S., and YAFFE, D. (1985). Developmentally regulated expression of a

Cell Line

121

chicken muscle-specific gene in stably transfected rat myogenic cells. Proc. Natl. Acad. Sci. USA 82, 3106-3109. ORDAHL, C. P. (1977). Reassociation kinetics of polyploid hepatocyte DNA. Biochim. Biophys. Acta 474, 17-29. ORDAHL, C. P., TILGHMAN, S. M., OVITT, C., FORNWALD, J., and LARGEN, M. T. (1980). Structure and developmental expression of the chick a-actin gene. Nucleic Acids Res. 8, 4989-5005. ORDAHL, C. P., EVANS, G. L., COOPER, T. A., KUNZ, G., and PERRIARD, J. C. (1984). Complete cDNA derived amino acid sequence of chick muscle creatine kinase. J. Biol. Chem. 259, 15,224-l&227. PINSET, C., and WHALEN, R. G. (1984). Manipulation of medium conditions and differentiation in the rat myogenic cell line L6. Des Biol. 102,269-277. PINSET, C., MONTARRAS, D., CHENEVERT, J., MINTY, A., BARTON, P., and LAURENT, C. (1988). Control of myogenesis in the mouse myogenie C2 cell line by medium composition and insulin: Characterization of permissive and inducible C2 myoblasts. Diferevtiation 38, 28-34. RICHLER, C., and YAFFE, D. (1970). The in vitro cultivation and differentiation capacities of myogenic cell lines. Deu. Biol. 23, l-22. RINALDI, L. M. (1959). An improved method for the isolation and quantitative cultivation of embryonic cells. E.rp. Cell Res. 16, 477505. SAWTELL, N. M., HARTMAN, A. L., and LESSARD, J. L. (1988). IJnique isoactins in the brush border of rat intestinal epithelial cells. Cell Motil. Cytoskel. 11, 318-325. SCHNEIDER, M. D., and OLSON, E. N. (1988). Control of myogenic differentiation by cellular oncogenes. Mol. Nwrobiol. 2, l-39. SCHUBERT, D., HARRIS, A. J., DEVINE, C. E., and HEINEMANN, S. (1974). Characterization of a unique muscle cell line. J. Cell Biol. 61, 398413. SPIZZ, G., ROMAN, D., STRAUSS, A., and OLSON, E. N. (1986). Serum and fibroblast growth factor inhibit myogenic differentiation through a mechanism dependent on protein synthesis and independent of cell proliferation. J. Biol. Chem. 261, 9483-9488. STERNBERG, E. A., SPIZZ, G., PERRY, W. M., VIZARD, D., WEIL, T., and OLSON, E. N. (1988). Identification of upstream and intragenic regulatory elements that confer cell-type-restricted and differentiationspecific expression on the muscle creatine kinase gene. Mol. Cell. Biol. 8, 2896-2909. STOCKDALE, F. E., and MILLER, J. B. (1987). The cellular basis of myosin heavy chain isoform expression during development of avian skeletal muscles. Dw. Biol. 123, l-9. TOYAMA, Y., FORRY-SCHAUDIES, S., HOFFMAN, B., and HOTZER, H. (1982). Effects of taxol and colcemid on myofibrillogenesis. Proc. Natl. Acad. Sci. USA 79, 6556-6560. WICK, M., and BANDMAN, E. (1989). Differentiating chicken skeletal muscle cells express a ventricular myosin heavy chain. J. Celell Biol. 109,262a. WRIGHT, W. E., SASSOON, D. E., and LIN, V. K. (1989). Myogenin, a factor regulating myogenesis, has a domain homologous to myoD. Cell 56, 607-617. YAFFE, D. (1968). Retention of differentiation potentialities during prolonged cultivation of myogenic cells. Proc. Natl. Acad. Sci. USA 61, 477-483. YAFFE, D., and SAXEL, 0. (1977a). A myogenic cell line with altered serum requirements for differentiation. Diflerentiation 7, 159-166. YAFFE, D., and SAXEL, 0. (1977b). Serial passaging and differentiation of myogenic cells isolated from dystrophic mouse muscle. Natuue (Londox) 270, 725-727.